Energy Management Strategy for Fuel Cell Hybrid Electric Vehicles Considering the Inaccuracy of Predicted Vehicle Speed
Huice Yang, Yunfeng Hu, Xun Gong, Ranhe Cao, Lulu Guo, Hong Chen
Abstract
Intelligent transportation creates opportunities for optimizing fuel cell hybrid electric vehicles (FCHEVs) energy. However, accurately predicting speeds is challenging for energy management. To address this problem, a model predictive control strategy considering (Con-MPC) vehicle speed inaccuracy is proposed. First, a Gaussian process (GP) is used to predict the vehicle speed with uncertainty. Second, under the MPC framework, the inaccuracy prediction is processed using a hierarchical structure. In the upper layer, the forward dynamic programming (FDP) is used to incorporate long-term inaccurate predictive information for solving the state of charge (SoC). The SoC is served as a reference and then transmitted to the lower layer at a frequency. In the lower layer, the Pontryagin minimum principle (PMP) is used to solve the optimization problem based on SoC guidance. Finally, the real-time implementation is evaluated in a dSPACE rapid prototyping system. The simulation results demonstrate that the Con-MPC strategy can enhance fuel economy by 1.7%-5.7% when compared to the basic MPC (Bas-MPC). Meanwhile, the improvement margin between Con-MPC and the benchmark is only 0.4%-10.93%. Furthermore, compared to the strategy that does not consider inaccurate vehicle speed, this strategy improves fuel economy by 1.11%.